Essays on religion as an industry

This dissertation is comprised of two chapters. The first chapter is titled "Estimating a Strategic Entry Model for Churches." This chapter treats the entry decisions of churches as if they were profit-maximizing firms and uses recent developments in the strategic entry industrial organization literature to study these decisions. A central theme in the entry literature is the potential for excess entry because the entrant fails to internalize the negative impact of its entry on the revenue of existing firms. Two key facts underlying my analysis are that Catholic churches tend to be much bigger in terms of members than Protestant churches and there are also fewer Catholic churches in a typical market. As compared to relatively decentralized Protestant churches, the Catholic Church is hierarchical, with authority for entry decisions vested in a local bishop. One might expect the bishop to internalize negative impact from entry of a new church, in a way that a Protestant preacher starting a new church would not. I estimate the parameters of an entry model using data from the entry of Protestant churches in specially defined markets and then do an experiment to determine how things look different when a Catholic bishop controls entry. I find that I can explain a large amount of the differences in entry patterns between Catholic churches and Protestant churches taking this difference in entry regulation into account.;The second chapter is titled, "A Model of Church Exit." Much work has been in the fields of economics and sociology treating religion as an industry. One significant empirical difference between churches and for-profit firms is that churches have much lower exit rates. This paper develops a model of the exit decisions of for-profit firms and religious entrepreneurs that differ in objectives, in what can be done with assets when a store or church (a unit) is shut-down and in the number of units they control. Historical data on the exit rates of churches and grocery stores support the predictions of the model.